Hydrocarbon mixtures such as gasoline and more recently diesel and jet fuels have served as transportation fuels for many years. Recent social and environmental concerns, particularly with respect to NO.sub.x and hydrocarbon gas emissions, however, have led to growing demand for alternate, cleaner burning fuels. Similarly, erosion of the ozone layer in the atmosphere and acid rain have led to a strong demand for controlled atmospheric emissions.
In this context, fuel oxygenates, particularly alcohols have emerged as strong contenders in the quest to develop cleaner burning fuels. Methanol is the cheapest and the most abundant of these alcohols. Favorable factors of methanol include its high octane rating, its manufacture from abundant natural resources (e.g. coal, gas, petroleum fractions, residual biomass and other agricultural products) and its ability to lessen environmental damage.
The current U.S. consumption of methanol is 1.3 billion gallons/yr., while the current U.S. gasoline consumption is 122 billion gallons/yr. Thus, it is likely that methanol will become a very important motor fuel or motor fuel supplement in the future.
Moreover, methanol can undergo a variety of reactions, some because of the presence of the hydroxide group, others because of the absence of steric hindrance of the methyl group, and still others because the --CH.sub.2 OH group is bound to the hydrogen atom rather than to another carbon atom. Methanol is, therefore, an important chemical precursor. It is widely used in the manufacture of formaldehyde (20% of total consumption), chloromethanes, acetic acid, methyl acetate and methyl formate. It is also used as an intermediate in the manufacture of acetic anhydride and in the manufacture of dimethyl ether. Methanol also finds increasing use as an octane booster for gasoline by direct blending or as a raw material for methyl tert-butyl ether (MTBE) and for fuel cell applications. Furthermore, there is the exciting discovery that methanol can be converted to high octane gasoline by the Mobil methanol-to-gasoline (MTG) process.
Methanol is currently produced almost solely by reacting synthesis gas ("syngas") comprising hydrogen and carbon monoxide in the presence of a heterogeneous copper catalyst. Methanol was first produced from syngas by Badische Anilin & Soda-Fabrik AG, Germany in 1923, according to the reaction shown in the following equation: EQU 2H.sub.2 +CO.fwdarw.CH.sub.30 H
Zinc/chromium oxide catalyst was used, with a high selectivity for formation of methanol at temperatures between 320 and 380.degree. C., and pressures of 300 to 350 atm. This catalyst along with minor variants was used in the "high pressure" methanol synthesis up to the mid 1960's. The reaction such as described in the above equation, in which methanol is produced from synthesis gas in one step, is often referred to as the "direct synthesis."
In 1966, a new "low pressure" process was developed by The Imperial Chemical Industries ("ICI"). This process uses Cu/ZnO or Cu/ZnO/Al.sub.2 O.sub.3 as a catalyst and operates at 200.degree.-300.degree. C. and 50-110 atm. Today, most methanol is manufactured by this method. The reaction is carried out in the gas phase in a fixed bed reactor. Approximately 6% by volume of CO.sub.2 is normally added to the syngas feed. It has been reported that all the methanol is formed via CO.sub.2 rather than CO as discussed by G. C. Chinchen, et al. in Chemtech, 692 (November 1990).
The power requirements, good catalyst life, larger capacity single-train convertor designs and improved reliability, of the low pressure technology result in lower energy consumption and economy of scale. However, the low pressure direct process has certain drawbacks. Chief among these drawbacks is the high operating temperature (T=250.degree. C.). Thermodynamic calculations show that, at a temperature of 250.degree. C. and a pressure of 50 atm, 51.9% of the syngas can be converted at equilibrium. These calculations are based upon a stoichiometric H.sub.2 /CO feed composition of 2 and an assumption of ideal gases. Under present industrial conditions, however, only about 6-12% conversion per pass is typically achieved.
The methanol synthesis reaction is very exothermic. Poor heat transfer in the catalyst bed results in an outlet methanol concentration limited to 5-6 mole%. Either cool unreacted gas injected at stages in the catalyst bed or internal cooling surfaces is generally used to control the bed temperature. To achieve maximum conversion of the carbon oxides, an excess of H.sub.2 is used. The excess of H.sub.2 requires a high recycle ratio which, in turn, leads to greater expense. Therefore, any modification in the process technology that can enhance heat transfer will result in higher conversions. Furthermore, a decrease in operating temperature could result in lower energy consumption and a higher equilibrium conversion.
To overcome the heat transfer limitation, slurry phase processes are being developed. Processes based on three phase fluidized bed, three phase fixed bed, slurry phase bubble column and mechanically agitated slurry phase reactors are known. These processes take advantage of the outstanding characteristics of a slurry reactor; notably the excellent heat transfer between the catalyst and the liquid, with the liquid serving as a heat sink for the heat of reaction. The use of slurry reactors results in excellent temperature control and higher synthesis gas conversion per pass. These processes, though not yet commercialized, operate at almost the same temperature and pressure as the gas phase process.
A process based on slurry phase technology is the "LaPorte process" being jointly developed by Air Products and Chem Systems (D. M. Brown and M. I. Greene, "Catalyst Performance in Liquid Phase Methanol Synthesis", presented at the Summer National AICLE Meeting, Philadelphia, August 1984). This process has been claimed to be near commercialization. The process development unit incorporates an ebullated slurry bubble column capable of once-through operation on clean coal gasifier effluent gas. The catalyst, which is suspended in the liquid phase, is a Cu/ZnO catalyst. Higher synthesis gas conversions per pass than achieved in gas-phase processes have been reported for the LaPorte process. The use of coal derived synthesis gas which is rich in CO has been reported to yield no substantial difference in the rate of methanol synthesis.
Although the use of slurry reactors provides beneficial results, several problems persist with the current "low pressure" technology. Most importantly, the synthesis temperature is still quite high. Because lower reaction temperatures result in lower free energy change for the reaction, the production of methanol is unfavorable at high temperatures. Temperatures of 240 to 260.degree. C. and high pressures are needed to achieve high reaction rates with the currently used process technology. Therefore, if catalysts with higher activity at lower temperatures could be developed, considerable improvement in the economics of the process would result.
A promising, but little studied, alternate route is a "two-step" synthesis to methanol via methyl formate disclosed in U.S. Pat. No. 1,302,011. The two-step synthesis comprises the carbonylation of a carrier alcohol to the corresponding alcohol formate using alkali alkoxides as homogeneous catalysts. The carbonylation step is followed by hydrogenolysis of the formate on the surface of copper chromite to yield the carrier alcohol and MeOH. The reaction sequence is shown by the following equations:
Carbonylation of carrier alcohol, EQU ROH+CO.fwdarw.HCOOR PA1 Hydrogenolysis of the corresponding formate, EQU ROOCH+2H.sub.2 .fwdarw.CH.sub.30 H+ROH
The two individual reactions are well known, the former being the commercial route to methyl formate production. The carbonylation reaction is carried out at temperatures of 80.degree.-100.degree. C. and pressures of 30-50 atm in the presence of a homogeneous catalyst. The carbonylation reaction thus takes place in the liquid phase. The hydrogenolysis of methyl formate can be carried out at temperatures of 100.degree.-160.degree. C. and atmospheric pressure in the presence of a heterogeneous catalyst. Carrying out these two steps in series can result in methanol synthesis at considerably milder conditions. If the carrier alcohol used is MeOH, the reaction yields two moles of MeOH as product. Reaction rates comparable to those obtained commercially and reduced separation costs are obtained by using methanol as the carrier alcohol.
The two-step process has several advantages over the direct methanol synthesis technology, including: lower reaction temperatures; higher synthesis gas conversions per pass, thus decreasing recycle load; and improved heat transfer, because the reaction is carried out in a liquid slurry.
The two-step synthesis process via alkyl formate avoids the thermodynamic limitations of presently practiced methanol synthesis, making the process less energy intensive. The liquid phase acts as a heat sink reducing the heat transfer limitation. When methanol is used as a solvent, mass transfer limitations are reduced. This process thus provides an efficient route to the manufacture of methanol.
A major disadvantage of the two-step process, however, is the need for two reactor systems and two feed preparation systems. A seemingly attractive alternative would be to carry out both reactions concurrently in the same reactor ("concurrent synthesis"). It is not clear, however, that such a combination of reactions would be feasible. Viewed independently, the carbonylation and hydrogenolysis reactions appear to be incompatible.
Initially, CO, one of the reactants in the carbonylation reaction, inhibits the hydrogenolysis reaction. Furthermore, CO.sub.2, which is usually present in syngas, has a strong negative effect on both reactions. The negative effect of CO.sub.2 in the carbonylation reaction appears to be irreversible when the CO.sub.2 is removed. Finally, selection of an operating temperature must be a compromise between the relatively low temperature required to obtain high conversion in the carbonylation reaction and the higher temperature required to obtain a reasonable rate in the hydrogenolysis reaction.
Experimental evidence showing that the carbonylation and hydrogenolysis reactions can occur in a single reactor containing sodium methoxide (NaOMe) at 200.degree. C. and 150-250 atm. and copper-chromium-calcium catalyst was provided by Imyanitov, N. B., et al., in Gidroformilirovanie, 152 (1972). Evidence showing that the concurrent reaction can occur at 200.degree. C. and 150-250 atm. with sodium carbonate or sodium formate in combination with a copper-chromium-calcium catalyst was also provided.
Aker Engineering, in Petrole Informations, May 13, 1982, also reported a two-component liquid phase catalytic system to convert syngas to a mixture of methanol and methyl formate in a single reactor. The process was reported to operate typically at 110.degree. C. and 0.5 MPa. Under these conditions the main product would be methylformate rather than methanol. The report disclosed the use of only alkali and/or alkaline earth alkoxides (alcoholates) as the carbonylation catalyst with copper chromite as the hydrogenolysis catalyst. The report also emphasized, however, the need to eliminate all CO.sub.2, H.sub.2 O and sulfur compounds from the inlet syngas.
Similarly, U.S. Pat. No. 4,731,386 discloses preparation of methanol from syngas in a liquid reaction mixture in the presence of a catalyst system consisting of an alkali alcoholate and a heterogeneous copper catalyst. It was found that the addition of a non-polar organic solvent having weak cation solvatizing properties in the liquid phase, otherwise consisting of methanol and methylformate, substantially increased the catalytic activity of catalyst systems consisting of an alkali metal. alcoholate and a heterogeneous copper catalyst.
Liu, Z., et al., "Methanol Synthesis via Methyl Formate in a Slurry Reactor", 18 Fuel Processing Technology, 185 (1988), studied concurrent synthesis in a slurry reaction using KOCH.sub.3 and a copper chromite at temperatures of 140.degree.-180.degree. C. and pressures of 3.8-6.2 MPa. Liu, et al. found the results from the concurrent methanol synthesis to be different from those predicted by the individual reactions. Methanol production was found to be higher and the effect of CO.sub.2 was lessened and reversible. A "small" amount of water was reported not to be detrimental to catalyst activity. Rates of reaction were found to be comparable to those reported for direct synthesis.
It is thus known that methoxides such as those derived from sodium or potassium or alkaline earth metals such as barium can be used along with a copper chromite catalyst in the concurrent synthesis of methanol. The main drawback of using a catalyst system including methoxides is the high cost required for catalyst manufacture and activity upkeep. Carbonates and formates have similarly been used in the concurrent process, but only at highly elevated temperatures.
The high temperatures currently needed to synthesize methanol via the direct route using copper-zinc catalysts and the high cost of using methoxides as a catalyst make it highly desirable to develop inexpensive alternative catalyst systems which enable methanol synthesis under mild temperatures while achieving high syngas conversion per pass and high methanol selectivity.